A student who is considering a Master’s degree in Statistics asks, “I’m interested in finding a job in data analysis and have been looking around, but I’m not sure if a masters is necessary to break into the field.”
Without much info about her background or job goals, here’s what I replied. Readers, do you have any additional or contradictory advice?
Do you have a clear sense of what you’d like to do with the degree? If it involves any kind of research work, my impression is that the credentials of a masters tend to be helpful for getting a job, especially with larger organizations that might have rigid hiring standards. My masters in statistics was certainly helpful in getting me a position at the Census Bureau. They do hire statisticians with a bachelors as well, but the masters gave me much more flexibility about which branch to work in and what projects to take on.
I don’t know your background, but if you haven’t taken a mathematical statistics course, a masters degree may be very useful in your work. There’s a huge difference between undergraduate Stats 101 (apply a few standard procedures to nice clean datasets) and real data analysis work (figure out how to clean the data and modify your procedures to the messy context in front of you). So a masters-level mathematical/theoretical stats course, where you learn to prove which estimators have desirable properties or to derive tests that are appropriate in a given situation, is invaluable when you run into non-standard problems. The masters degree will also expose you to many techniques that you probably didn’t cover as an undergrad: designing good experiments, computer-intensive methods like the bootstrap, special-use techniques like time series or spatial statistics, other inference philosophies like Bayesian statistics, etc. Finally, if your program requires any consulting and/or a thesis, these are useful concrete projects to have on your resume and bring up in job interviews.
On the other hand, depending on your background and job goals, you might not need the masters. Some “data analysis” jobs require nothing more than basic spreadsheet skills (keeping track of the office’s paper supplies, combining monthly sales reports from the company’s branches, etc).
Then there are the data-related jobs such as “data-driven journalism” that also require some programming skills (scraping data from the web, mashing up different databases, designing and implementing beautiful interactive visualizations, etc). These skills are sadly not taught in most statistics departments. My stats-programming coursework covered how to use and develop statistical procedures assuming you have clean data, but not how to get things into the right format or how to show off my results online.
Now, some of the most exciting opportunities out there are for someone who has both this developer skillset and a deep knowledge of statistics. But while the coding can be self-taught through trial and error, I think the statistics is still best learned in a rigorous program. If my web graphic doesn’t display, I know there must be a bug; but just because my statistical routine ran and produced some numbers doesn’t mean it produced the right numbers or that I’ve interpreted them correctly. And likewise, an impressive portfolio of your self-taught coding work might be enough to land you programming jobs, but it can be harder for a self-taught statistician to show potential employers that they know their stuff.
After she told me a bit more about her background, a followup:
If you’ve already worked as a financial analyst for a few years, you might already have the skills you need for the jobs you want. I will say that for the project you mentioned (about making government groups more efficient), some programming skills in R or Python will certainly help you process the datasets more efficiently than with things like Excel. Also, I still think masters-level stats coursework would be useful on a project like that, but maybe it’s not necessary right from the start. If you don’t do the MS now, you could also consider taking some advanced stat courses in the evenings as you’re earning a day-job salary, instead of living as a poor grad student all year 🙂
If you want to teach yourself R, here are the materials from a course last year here at CMU. Also check out the R textbooks listed there.